Legal claims defining the scope of protection. Each claim is shown in both the original legal language and a plain English translation.
1. A method of computing a time-to-contact (TTC) of a vehicle with an object in an environment of the vehicle, wherein the object includes a light source, the method performable by a camera connectible to a processor on-board the vehicle, the method comprising: capturing, via the camera, a plurality of images of the object; tracking, by the processor, a spot identified in at least two of the plurality of images, wherein the tracked spot includes light emanating from a light source of another vehicle; determining, by the processor, a change in brightness of the tracked spot based on a change in respective pixel values of the tracked spot in the at least two of the plurality of images; determining, by the processor, a change in computed energy of the tracked spot based on a change in aggregate pixel values in at least one image of the at least two of the plurality of images; and computing, by the processor, the time-to-contact based on the change in brightness of the tracked spot and the change in computed energy of the tracked spot.
2. The method according to claim 1 , wherein the energy is computed by summing pixel values from a plurality of pixels of the tracked spot.
3. The method according to claim 1 , further comprising: fitting a function of the energy to a function of the time-to-contact.
This invention relates to a method for analyzing energy and time-to-contact in a system, particularly for applications in collision prediction, robotics, or autonomous navigation. The method addresses the challenge of accurately estimating the remaining time before a collision or interaction occurs based on energy measurements, which is critical for real-time decision-making in dynamic environments. The method involves measuring energy over time and determining a time-to-contact value, which represents the estimated time until a collision or interaction. The core innovation is the fitting of a function of the energy to a function of the time-to-contact. This step establishes a mathematical relationship between the energy dynamics and the time-to-contact, allowing for precise predictions. The fitting process may involve regression analysis, curve fitting, or other mathematical techniques to model the relationship accurately. Additionally, the method includes preprocessing the energy data to remove noise or irrelevant variations, ensuring that the fitted function is based on clean, reliable measurements. The time-to-contact function may be derived from physical principles, empirical data, or a combination of both, depending on the application. The method may also incorporate adaptive adjustments to the fitting parameters to improve accuracy over time or under varying conditions. By establishing a direct relationship between energy and time-to-contact, this method enables systems to anticipate collisions or interactions with high precision, improving safety and efficiency in applications such as autonomous vehicles, industrial robotics, and collision avoidance systems.
4. The method of claim 3 , wherein a reciprocal of a square root of the energy is fit to a linear function of the time-to-contact.
A method for analyzing motion data involves determining a time-to-contact (TTC) between a moving object and a reference point. The method calculates an energy value associated with the motion of the object, then computes the reciprocal of the square root of this energy. This computed value is then fit to a linear function of the time-to-contact. The linear function provides a predictive model for estimating the TTC based on the energy dynamics of the object's motion. This approach is particularly useful in applications where precise timing of an object's arrival at a reference point is critical, such as in collision avoidance systems, robotics, or autonomous navigation. The method leverages the relationship between energy dissipation or accumulation and the remaining time before contact, offering a computationally efficient way to predict TTC without requiring complex motion tracking or sensor data. The linear fitting process allows for real-time adjustments and improvements in accuracy as new data is collected. This technique can be applied to various types of motion, including linear and non-linear trajectories, and is adaptable to different environmental conditions. The method ensures reliable TTC predictions by continuously refining the linear model based on observed energy changes over time.
5. The method according to claim 1 , further comprising: determining whether the vehicle and the object are on a collision course responsive to detecting motion of the tracked spot in the plurality of images.
This invention relates to collision detection systems for vehicles, specifically methods for determining whether a vehicle and an object are on a collision course. The system tracks a spot on an object in a series of images captured by a vehicle-mounted camera. Motion of the tracked spot is analyzed to assess whether the vehicle and the object are moving toward each other in a way that would result in a collision. The method involves capturing multiple images of the object, identifying a spot on the object in each image, and calculating the motion of the spot between images. If the motion indicates the object is moving toward the vehicle, the system determines a collision course exists. This approach improves collision detection accuracy by dynamically tracking object movement rather than relying on static position data. The system may also adjust tracking parameters based on environmental conditions, such as lighting or distance, to maintain reliable detection. The invention is particularly useful for autonomous vehicles or advanced driver-assistance systems (ADAS) that require precise collision prediction to trigger avoidance maneuvers or warnings. By continuously monitoring object motion, the system enhances safety by reducing false positives and ensuring timely collision alerts.
6. The method of claim 5 , wherein determining whether the vehicle and the object are on a collision course includes a determination with respect to an amount of time before the spot moves a predetermined vertical distance in the plurality of images as compared to an amount of time before the spot moves a predetermined horizontal distance in the plurality of images.
This invention relates to collision detection systems for vehicles, specifically methods for determining whether a vehicle and an object are on a collision course by analyzing motion patterns in a sequence of images. The problem addressed is the need for accurate and efficient collision prediction to enhance vehicle safety, particularly in autonomous or semi-autonomous driving systems. The method involves tracking a spot representing an object in a series of images captured by a vehicle-mounted camera. The system determines whether the vehicle and the object are on a collision course by comparing the time it takes for the spot to move a predetermined vertical distance in the images to the time it takes for the spot to move a predetermined horizontal distance. If the vertical and horizontal motion times meet specific criteria, the system concludes that a collision is likely. This approach leverages the relative motion of the object in the image plane to infer real-world collision risk without requiring complex 3D reconstruction or depth sensing. The method may also involve preprocessing the images to enhance object detection, such as applying filters or thresholding techniques to isolate the spot. The system may further adjust the predetermined vertical and horizontal distances based on environmental factors like lighting conditions or object size. By focusing on motion analysis in the image plane, the method provides a computationally efficient way to assess collision risk, reducing the need for high-resolution sensors or extensive processing power. This technique is particularly useful in real-time applications where rapid decision-making is critical for avoiding accidents.
7. The method of claim 1 , wherein the energy is computed from a predicted spot intensity based on non-saturated pixels at an edge of the tracked spot.
A method for computing energy in optical tracking systems addresses the challenge of accurately determining spot intensity in imaging applications where saturation occurs. The technique focuses on non-saturated pixels at the edge of a tracked spot to predict spot intensity, enabling precise energy calculations even when central pixels are saturated. This approach improves reliability in high-intensity scenarios where traditional methods fail due to saturation artifacts. The method involves analyzing pixel data at the spot's periphery, where saturation is less likely, to estimate the full spot intensity. By leveraging these non-saturated edge pixels, the system avoids errors caused by clipped intensity values, ensuring accurate energy computation. This technique is particularly useful in applications like laser tracking, optical sensing, and imaging systems where precise intensity measurements are critical. The method enhances performance in environments with varying light conditions, providing consistent and reliable energy calculations for improved tracking accuracy.
8. The method of claim 1 , wherein the change in the energy over time is determined based on a single frame value, wherein the single frame value is calculated from a histogram of energy changes between a current compared image frame and multiple previous image frames, and wherein the energy changes are weighted in the histogram based on a relative lapse in time between the current compared image frame and the multiple previous image frames.
This invention relates to image processing, specifically to methods for analyzing energy changes between image frames to detect motion or other dynamic events. The problem addressed is the need for accurate and efficient motion detection in video sequences, particularly where changes in energy over time must be precisely measured to distinguish relevant events from noise. The method involves calculating a single frame value representing energy changes between a current image frame and multiple previous frames. This value is derived from a histogram of energy changes, where each entry in the histogram corresponds to the difference in energy between the current frame and a prior frame. The histogram entries are weighted based on the time elapsed between the current frame and each previous frame, ensuring that more recent frames contribute more significantly to the energy change calculation. This weighting helps prioritize recent motion while accounting for historical context, improving detection accuracy. The technique is particularly useful in applications like surveillance, video compression, and automated event detection, where distinguishing meaningful motion from background noise is critical. By using a weighted histogram approach, the method balances responsiveness to recent changes with stability against transient fluctuations, enhancing reliability in dynamic environments.
9. The method of claim 8 , wherein the change in the energy over time is determined based on a weighted sum of a current single frame value and a plurality of prior single frame values calculated from a plurality of prior current compared image frames.
This invention relates to image processing, specifically to methods for analyzing changes in energy over time in a sequence of image frames. The problem addressed is accurately tracking dynamic changes in visual data, such as motion or intensity variations, by incorporating both current and historical frame data to improve stability and accuracy in energy calculations. The method involves determining a change in energy over time by computing a weighted sum of a current single frame value and multiple prior single frame values. These prior values are derived from comparisons between a current image frame and multiple preceding image frames. The weighted sum approach allows for smoothing and noise reduction while preserving temporal dynamics, making it useful for applications like motion detection, video compression, or object tracking. The method ensures that the energy calculation is not solely dependent on the most recent frame but also considers historical data, which helps mitigate transient noise and improves robustness. The weights applied to the current and prior values can be adjusted based on application requirements, such as prioritizing recent changes or maintaining long-term stability. This approach enhances the reliability of energy-based analysis in dynamic environments.
10. The method of claim 1 , wherein tracking the spot identified in at least two of the plurality of images includes determining whether fluctuations in energy for the tracked spot between images correspond to known fluctuations associated with a known type of light source.
A method for analyzing light sources in image sequences involves tracking a spot of light across multiple images to determine whether the light source is of a known type. The method includes capturing a sequence of images containing the light spot, identifying the spot in at least two of the images, and tracking its position over time. The tracking process involves measuring fluctuations in the spot's energy (e.g., brightness or intensity) between images and comparing these fluctuations to known patterns associated with specific light sources, such as artificial lighting, natural light, or other identifiable sources. By matching the observed fluctuations to predefined energy fluctuation profiles, the method can classify the light source type. This approach is useful in applications like surveillance, astronomy, or automated lighting control, where distinguishing between different light sources is important. The method may also involve preprocessing the images to enhance spot detection, such as filtering noise or adjusting contrast, to improve tracking accuracy. The comparison step may use statistical analysis or machine learning to determine the closest match between observed and known fluctuation patterns.
11. A system for computing a time-to-contact (TTC) of a vehicle with an object in an environment of the vehicle, wherein the object includes a light source, the system mountable in the vehicle, the system comprising: a camera; and a processor configured to: capture from the camera a plurality of images of the object; track a spot identified in at least two of the plurality of images, wherein the tracked spot includes light emanating from a light source of another vehicle; determine a change in brightness of the tracked spot based on a change in respective pixel values of the tracked spot in the at least two of the plurality of images; determine a change in computed energy of the tracked spot based on a change in aggregate pixel values in at least one image of the at least two of the plurality of images; and compute the time-to-contact (TTC) based on the change in brightness of the tracked spot and the change in computed energy of the tracked spot.
The system computes the time-to-contact (TTC) between a vehicle and an object, such as another vehicle, by analyzing light sources in the environment. The system includes a camera and a processor. The camera captures multiple images of the object, which contains a light source. The processor tracks a specific light spot from at least two of these images, where the spot originates from the light source of another vehicle. The processor then determines the change in brightness of the tracked spot by comparing pixel values in the images. Additionally, the processor calculates the change in computed energy of the spot by analyzing aggregate pixel values in at least one of the images. Using these changes in brightness and energy, the processor computes the TTC, providing an estimate of when the vehicle will collide with the object. This approach leverages visual data to enhance collision prediction accuracy, particularly in scenarios where traditional sensors may be limited. The system is designed to be mounted in the vehicle and operates by continuously processing image data to assess collision risks dynamically.
12. The system according to claim 11 , wherein the energy is computed by summing pixel values from a plurality of pixels of the tracked spot.
The system relates to energy computation in imaging or tracking systems, particularly for analyzing a tracked spot within an image. The problem addressed involves accurately determining the energy or intensity of a tracked spot, which is essential for applications such as particle tracking, medical imaging, or industrial inspection. Traditional methods may struggle with noise, varying pixel values, or computational efficiency when calculating energy from a tracked spot. The system computes the energy of a tracked spot by summing pixel values from multiple pixels within the spot. This approach improves accuracy by aggregating contributions from all relevant pixels rather than relying on a single measurement. The tracked spot is identified within an image, and the pixel values within its boundaries are summed to derive the total energy. This method ensures robustness against noise and variations in pixel intensity, providing a more reliable energy measurement. The system may also include preprocessing steps, such as filtering or normalization, to enhance the accuracy of the energy computation. The summed pixel values can be used for further analysis, such as tracking motion, detecting anomalies, or quantifying intensity changes over time. This technique is particularly useful in high-resolution imaging systems where precise energy measurements are critical.
13. The system according to claim 11 , wherein the processor is further configured to: fit a function of the energy to a function of the time.
This invention relates to a system for analyzing energy data over time, particularly in applications where understanding the relationship between energy and time is critical, such as in energy management, power grid optimization, or industrial process monitoring. The problem addressed is the need to accurately model and predict energy behavior by fitting a mathematical function to energy data as it varies over time. This allows for better decision-making in energy distribution, consumption forecasting, and system efficiency improvements. The system includes a processor that processes energy data collected over time. The processor is configured to fit a function of energy to a function of time, meaning it applies mathematical modeling techniques to derive a relationship between energy values and their corresponding time points. This involves selecting an appropriate function type (e.g., linear, polynomial, exponential) and optimizing parameters to minimize the difference between the model and the observed data. The fitted function can then be used to predict future energy values, detect anomalies, or optimize energy usage patterns. The system may also include data acquisition components to collect energy measurements and storage for historical data, enabling long-term trend analysis. By dynamically fitting energy data to time-based functions, the system provides a more accurate and adaptable model for energy analysis compared to static or rule-based approaches.
14. The system according to claim 11 , wherein the camera is a camera from a stereo pair of cameras.
A system for capturing and processing visual data includes a stereo pair of cameras configured to capture overlapping images of a scene from different perspectives. The stereo cameras enable depth perception and three-dimensional reconstruction of the scene by analyzing disparities between the images. The system processes the captured images to generate depth maps, point clouds, or other 3D representations, which can be used for applications such as object detection, scene reconstruction, or augmented reality. The stereo configuration allows for accurate depth estimation by triangulating corresponding points between the two camera views. The system may further include image processing algorithms to enhance image quality, reduce noise, or correct distortions, ensuring reliable depth information. The stereo cameras are synchronized to capture images simultaneously or in rapid succession, minimizing motion artifacts. The system may also incorporate calibration techniques to align the cameras and ensure consistent depth measurements. This approach improves spatial awareness in applications requiring precise 3D modeling or environmental mapping.
15. The system according to claim 11 , wherein the time-to-contact is computed from a sensor input to the processor from a combination of the camera with a radar system.
A system for computing time-to-contact in autonomous vehicle navigation uses sensor fusion to improve accuracy. The system combines data from a camera and a radar system to determine the time remaining before a vehicle collides with an obstacle. The camera captures visual information about the environment, while the radar system provides distance and velocity measurements. By integrating these inputs, the processor calculates a more reliable time-to-contact estimate than either sensor could provide alone. This approach reduces errors caused by environmental factors like lighting conditions or sensor noise. The system is particularly useful for collision avoidance, adaptive cruise control, and emergency braking applications. The fusion of camera and radar data enhances situational awareness, allowing the vehicle to react more effectively to dynamic obstacles. This method improves safety by providing a more precise prediction of potential collisions, enabling timely corrective actions. The system can be integrated into existing autonomous driving platforms to enhance their performance in real-world scenarios.
16. The system according to claim 11 , wherein the time-to-contact is computed from a sensor input to the processor from a combination of the camera with a lidar system.
The system relates to computing time-to-contact in autonomous vehicle navigation, addressing the challenge of accurately determining the time remaining before a vehicle collides with an obstacle. The system integrates sensor data from a camera and a lidar system to enhance the precision of time-to-contact calculations. The camera captures visual information, while the lidar system provides depth and distance measurements. The processor analyzes the combined data to estimate the time until contact with an object, improving collision avoidance and navigation decisions. The lidar system supplements the camera by providing high-resolution spatial data, reducing errors from visual occlusions or lighting variations. This hybrid approach ensures reliable time-to-contact predictions in diverse environmental conditions, supporting safer autonomous driving. The system may also include additional sensors or processing modules to refine the calculations further. The integration of lidar with camera data enables real-time, accurate distance and velocity assessments, critical for dynamic obstacle detection and avoidance in autonomous systems.
Unknown
December 8, 2020
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.